Toward a standard formal semantic representation of the model card report.
Toward a standard formal semantic representation of the model card report.
复制标题
迈向模型卡报告的标准正式语义表示。
DOI：
10.1186/s12859-022-04797-6
复制
发表时间：
2022-07-14
影响因子：
3
通讯作者：
Tao, Cui
中科院分区：
文献类型：
--
作者：
Amith, Muhammad Tuan;Cui, Licong;Zhi, Degui;Roberts, Kirk;Jiang, Xiaoqian;Li, Fang;Yu, Evan;Tao, Cui
关键词：
模型卡报告旨在为利益相关者提供机器学习模型的信息和透明描述。本报告文档对美国国立卫生研究院的Bridge2AI计划感兴趣，该计划旨在通过基于人工智能的机器学习模型解决FAIR在生物医学研究中的挑战。我们提出了我们早期的承诺，在开发一个本体捕捉嵌入在模型卡报告的概念级信息。在已有本体的基础上，开发了模型卡报表本体的核心框架，生成了模型卡报表本体。我们的开发工作产生了一个基于OWL2的工件，它表示并形式化了模型卡报告信息。此本体的当前版本利用了OBO Foundry本体的标准概念和属性。此外，软件推理器表明与本体没有逻辑上的不一致。通过生物信息学研究（HIV社交网络和支架植入的不良结局预测）的机器学习模型的样本模型卡，我们展示了我们的模型在将静态模型卡报告转换为基于机器的处理的可计算格式方面的覆盖范围和有用性。我们的工作的好处是，它利用了广泛的和标准的术语和科学严谨的生物医学本体学家，以及，生成一个途径，使模型卡机器可读的语义网技术。我们未来的目标是评估我们的模型的准确性，并在以后扩展模型，以包括更多的概念，以解决术语的差距。我们讨论的工具和软件，将利用我们的本体为潜在的应用服务。
Model card reports aim to provide informative and transparent description of machine learning models to stakeholders. This report document is of interest to the National Institutes of Health’s Bridge2AI initiative to address the FAIR challenges with artificial intelligence-based machine learning models for biomedical research. We present our early undertaking in developing an ontology for capturing the conceptual-level information embedded in model card reports. Sourcing from existing ontologies and developing the core framework, we generated the Model Card Report Ontology. Our development efforts yielded an OWL2-based artifact that represents and formalizes model card report information. The current release of this ontology utilizes standard concepts and properties from OBO Foundry ontologies. Also, the software reasoner indicated no logical inconsistencies with the ontology. With sample model cards of machine learning models for bioinformatics research (HIV social networks and adverse outcome prediction for stent implantation), we showed the coverage and usefulness of our model in transforming static model card reports to a computable format for machine-based processing. The benefit of our work is that it utilizes expansive and standard terminologies and scientific rigor promoted by biomedical ontologists, as well as, generating an avenue to make model cards machine-readable using semantic web technology. Our future goal is to assess the veracity of our model and later expand the model to include additional concepts to address terminological gaps. We discuss tools and software that will utilize our ontology for potential application services.
登录
查看更多内容
DOI：
10.1007/s10817-014-9305-1
发表时间：
2014-10-01
期刊：
JOURNAL OF AUTOMATED REASONING
影响因子：
--
作者：
Glimm, Birte;Horrocks, Ian;Wang, Zhe
通讯作者：
Wang, Zhe
影响因子：
2.5
作者：
Grau, Bernardo Cuenca;Horrocks, Ian;Sattler, Ulrike
通讯作者：
Sattler, Ulrike
影响因子：
1.9
作者：
Malone, James;Brown, Andy;Stevens, Robert
通讯作者：
Stevens, Robert
影响因子：
3
作者：
Horridge, Matthew;Bechhofer, Sean
通讯作者：
Bechhofer, Sean
影响因子：
3
作者：
Berners-Lee, T;Hendler, J;Lassila, O
通讯作者：
Lassila, O